Research and development of mineral identification utilizing hyperspectral remote sensing

This paper reviews the current progress of mineral identification based on hyperspectral remote sensing. The physicochemical mechanisms of mineral spectra and their feature measurement and analysis are introduced. The properties of current and future main hyperspectral sensors are concluded. Then, three series of methods (based on spectral absorption feature, full spectral profile matching, and spectral unmixing, respectively) for mineral identification based on hyperspectral data sets are comparatively analyzed and summarized. Finally, the main problems of mineral identification by using hyperspectral data on theory, data sets, methods and applications, are analyzed and its developing trends are discussed. It points out that the trend will focus on the direction from the qualitative identification to the quantitative mineral analysis. During this trend, spectral unmixing, the design of new hyperspectral sensors for mineral identification, intellectualization of mineral information extraction, and mineral identification in a complex geological environment will be the main research focuses.